GLM
Z.ai's open-weight family, behind the cheap coding subscriptions.
GLM is the model family from Z.ai, and the reason its coding plan turns up in every "cheaper than Claude Code" thread — the weights are published, plenty of providers serve them, and the Flash models are some of the cheapest usable coding models anywhere. The 5.3 line runs to a million tokens of context, with Flash variants that also read images and video. Z.ai's own chat app is listed separately under Chats.
At a glance — GLM 5.2
- Cost
- $0.65 in · $2.042 out per 1M tokensMid-range
- Memory
- 1M tokens — about 1,550 pages
- Reads
- Text
- Can do
- Use tools (agents)
- Structured output
- Thinking mode
- Good at
- Coding: better than 72% · Agentic work: better than 72% · Overall: better than 61%
- Reliability
- 99.99% uptime, last 24h
- Released
- Jun 16, 2026
Advanced details
- API model ID
z-ai/glm-5.2- Max output
- 131K tokens
- Max input
- 1M tokens
- Cached input
- $0.121 per 1M
- Tokenizer
- Other
- Coding index
- 68.8
- Agentic work index
- 38.4
- Overall index
- 33.7
Providers
The same model, hosted by different companies. OpenRouter picks one per request and falls back to the next if it fails.
| Provider | Input /1M | Output /1M | Cached /1M | Max output | Uptime 24h |
|---|---|---|---|---|---|
| Baidu · fp8 | $0.489 | $1.536 | $0.091 | 131K | 99.93% |
| Inceptron · fp4Degraded | $0.489 | $2.073 | $0.182 | 944K | 98.48% |
| Decart · mxfp4 | $0.49 | $1.54 | $0.091 | 118K | 99.77% |
| StreamLake · fp8 | $0.556 | $1.746 | $0.103 | 128K | 99.70% |
| DeepInfra · fp4 | $0.563 | $1.80 | $0.105 | 164K | 99.39% |
| Baidu · fast | $0.648 | $2.269 | $0.161 | 131K | 99.99% |
| Novita · fp8 | $0.65 | $2.042 | $0.121 | 131K | 99.87% |
| DigitalOcean | $0.70 | $2.20 | $0.105 | 236K | 99.89% |
| CoreWeave · fp4 | $0.76 | $2.42 | $0.14 | 944K | 98.01% |
| AtlasCloud · fp8 | $0.938 | $2.948 | $0.174 | 131K | 99.21% |
| Alibaba · fp8 | $0.966 | $3.036 | $0.193 | 131K | 99.43% |
| SiliconFlow · fp8 | $1.19 | $3.74 | $0.221 | 262K | 86.74% |
| Phala · fp8 | $1.26 | $3.00 | $0.22 | 131K | 99.45% |
| BaseTen · fp8 | $1.40 | $4.40 | $0.14 | 262K | 99.96% |
| Together | $1.40 | $4.40 | $0.26 | 944K | 99.58% |
| Wafer | $1.40 | $4.40 | $0.26 | 944K | 99.35% |
| Fireworks | $1.40 | $4.40 | $0.14 | 944K | 99.97% |
| BaseTen · fp8 | $1.40 | $4.40 | $0.14 | 262K | 99.89% |
| Venice · fp8 | $1.40 | $4.40 | $0.26 | 131K | 99.89% |
| GMICloud · fp8 | $1.40 | $4.40 | $0.26 | 944K | 99.18% |
| Parasail · fp4 | $1.40 | $4.40 | $0.26 | 236K | 99.93% |
| Friendli | $1.40 | $4.40 | $0.26 | 944K | 99.81% |
| Cloudflare | $1.40 | $4.40 | $0.26 | 236K | 99.88% |
| Z.AI · fp8 | $1.40 | $4.40 | $0.26 | 131K | 99.68% |
| BaseTen · fast | $2.10 | $6.60 | $0.21 | 262K | 97.83% |
| Fireworks · fast-us | $2.10 | $6.60 | $0.21 | 944K | 99.84% |
| BaseTen · fast | $2.10 | $6.60 | $0.21 | 262K | 98.83% |
| Fireworks · fast | $2.10 | $6.60 | $0.21 | 944K | 99.51% |
| Decart · fast | $2.25 | $8.00 | $0.48 | 944K | 99.89% |
| Alibaba · fast | $2.31 | $7.26 | $0.462 | 131K | 97.56% |
Supported parameters
- frequency_penalty
- include_reasoning
- logit_bias
- logprobs
- max_tokens
- min_p
- parallel_tool_calls
- presence_penalty
- reasoning
- reasoning_effort
- repetition_penalty
- response_format
- seed
- stop
- structured_outputs
- temperature
- tool_choice
- tools
- top_k
- top_logprobs
- top_p
Live from OpenRouter, refreshed hourly. Scores from Artificial Analysis.
Key info
- Pricing
- Open source
- Category
- Models
- Best for
- Intermediate
- Made by
- Z.ai
- Weights
- Open
- Good for
- Coding on a budget